Writing the methods section of your grant

Steve Simon

2001-01-15

Dear Professor Mean, I’m starting to write the methods section for a research grant, but I have no idea where to start. – Dazed Dana

Dear Dazed,

Just explain that any moron could do the research you are proposing, and that makes you well qualified to do this research.

Short explanation

The methods section of a grant describes the process that you will follow to conduct your research. You should emphasize the complex, unusual, or innovative aspects of your research. Provide enough details so that someone reading this section could conduct or replicate your research.

More details

Here’s a list of elements to consider when writing a methods section. Every research project is different, so you will not necessarily use all of these elements. There are several elements that should be described in a methods section:

Also be sure to document any unusual materials or methods that are part of your research.

What are you measuring?

Research involves measurement. There are three basic types of measurements:

Outcomes measurements are those values that you are trying to influence or predict. These measurements represent the yardstick by which you will assess the success of a new therapy or the danger of a toxic exposure. Think of outcomes as effects in a hypothesized cause-and-effect relationship.

Independent variables are those values which you believe may influence the outcome measures and which are of direct interest. Independent variables might represent variables measured on a continuum like blood pressure. They also might represent groups such as treatment versus placebo, or exposure versus control. Independent variables will usually be those variables that are of sufficient research interest that you specified in your research hypotheses.

Covariates are also variables which you believe may influence the outcome measure, but are those which are not of direct research interest. Think of these as nuisance variables that you have to account for as part of the research. In many contexts, demographic variables like age, race, and gender are covariates. Be careful, though, because a demographic factor could be of direct research interest, as in a study of how socioeconomic factors influence levels of prenatal care.

For any of the above measurements that involve a level of subjectivity or judgement, you need to document who will be making those measurements. Also be sure to describe any efforts you take to ensure the quality of those measurements (e.g., independent assessment by a second examiner).

Whom are your measuring?

If your research directly involves human subjects, you need to specify how they were selected and assigned to groups. Document any payments they might receive and any agreements that they might sign. Be sure to specify to what population these results will be generalized.

If your research involves animals, specify the genus, species, and strain numbers. Also describe their housing conditions.

If your research involves the sampling of materials or equipment (e.g., home pregnancy tests), be sure to specify where and how you purchase this equipment.

When are you measuring?

You should document the timing of your measurements. Must they occur at a certain stage of life (e.g., during the first 24 hours of life) or at a certain time of the day (e.g., immediately after awaking in the morning)? If you are intervening, define when the measurements are done relative to the intervention.

Specifying the timing is especially important in a pre-test/post-test research design and in a longitudinal study. Both designs are set up to measure and evaluate changes. You need to specify whether you are looking at short term or long term changes, and explain why.

How are you intervening?

Not all research involves direct intervention with the subjects. If yours does, you need to document this intervention. Explain the instructions given to the research subjects? What information will be hidden from the subjects (and those evaluating the subjects) during the course of the research. How does the intervention differ between the treatment and control group?

Unusual materials or methods

Provide details for any specialized equipment. This would include the vendor name and location, and any special calibration procedures.

Also describe any specialized procedures. Explain how they will be performed and who is allowed to perform them.

Describe any survey (or questionnaire) that you use. Cite the source of the survey, and any information about the quality of this survey (quality measures that you have conducted with pilot data and/or measures from previous publications using this survey). Estimate how much time the typical subject will need for the survey, and any special accommodations for language or literacy limitations. Explain how you will handle non-response on important survey items and/or failure to return the survey.

You do not need to document routine equipment or procedures. Most health professionals know the method for taking a rectal temperature, and those of us who are not health professionals don’t want or need to know.

Summary

When you are outlining the research methods in a protocol, be sure to specify what measurements you are making, on whom you are making them, when you are measuring, and how you are intervening. Also be sure to document any unusual materials or methods.

Further reading

While the Lang and Secic book described below discusses writing a publication rather than a protocol, there is much in common between the two. Pay special attention to pages 6-21.

How to Report Statistics in Medicine. Lang TA, Secic M. Phildelphia PA: American College of Physicians (1997). ISBN: 0-943126-44-4.

You can find an earlier version of this page on my original website.